It’s a very common complaint that sites that focus on measuring the frequency reponses of DACs, amplifiers, speakers, and headphones don’t fully capture the capabilities of human hearing. This goes to the heart of the “subjective vs. objective” debate with subjectivists arguing that only by listening to equipment can audiophiles really understand if it is good or bad for their particular setup and needs. Objectivists argue that proper measurements give a more, well, objective understanding of the quality of the music reproduction of one device compared to others.
In subjective descriptions of equipment there are common adjectives and phrases that are used to qualify the sound characteristics of audio devices like “buttery,” “euphonic,” “wide soundstage,” “good spatial characteristics,” “muddy bass,” and so forth.
Frequency response measurements use sophisticated devices like Audio Precision analyzers that produce reference signals and then perform Fast Fourier Transforms to establish how well the input signal is reproduced at the output. That is what we mean by “high fidelity” in general: the device should transparently reproduce the input. Single frequencies are not the only test, however. An Intermodulation Distortion (IMD) test uses groups of tones simulaneously to examine whether the reproduction of them results in interference between the tones. After all, music isn’t just a single sine wave. Frequency sweeps from 20Hz to 20kHz or higher provide additional insights.
For measuring speakers, there are reference analyzers that rotate a microphone around the speaker on a robotic arm, testing the near- and far-field characteristics of the speaker and combining the recordings together mathematically to establish a better-than-anechoic model of the speaker’s output response. (Anechoic chambers are these crazy rooms filled with foam spikes that absorb all the reflected sounds so measurements can be reliably taken.)
Likewise, headphones can be evaluated using sophisticated measurement fixtures that even simulate human ear structure.
Is there any truth to the idea that measurements don’t tell the whole story? Sometimes, yes! But often the subjectivists are simply mixing things up. For example, we know from both measurements and listening that most tube or valve amplifiers tend to add in harmonic distortion when reproducing music. Some people like that distortion and call it “euphonic.” Vinyl record reproduction has limits on the fidelity that can be achieved using vinyl as a medium but many like the sound that vinyl equipment chains render. That’s perfectly OK, of course. It’s OK to have preferences, nostalgia, and opinions.
Where a mistake is often made, however, is in assuming that because the measurements show the flaws in noise and distortion that those measurement are insufficient, wrong, or bad. They aren’t. The measurements show objective insights into the performance of the device and the limitations in the fidelity of the reproduction. They also can help understand how to adjust the signal using equalization and digital signal processing in order to overcome the limitations of the device itself.
Note that the current cross-section of standard measurement technologies and techniques are not the end of the story, though. There are some interesting tests and arguments covered elsewhere that suggest some slight improvements might be possible. Science and engineering are evolutionary in their own way and knowledge production continues apace.
Are there limits in using tools like these analyzers to establish the fidelity of equipment? Of course, but it should be noted that engineers use exactly the same technologies in their design process. For instance, mathematical modeling might be used for a speaker case design as well as for the time and frequency behavior of drivers, but often a Klippel system will be used to confirm that the design fulfills expectations about performance. But some of the assumptions inherent in those measurement tools might be violated for electrostatic speakers resulting in difficulty in correlating the experience of listening to the speaker with the measurements. Here, preferences play a greater role than measurements in establishing the value of the speaker to the listener and the limits of the measurement setup come into play.
So a sensible balance can be maintained about the value of measurements while still acknowledging that they provide an essential ground truth when done correctly and interpreted smartly.
